The catabolite gene activator protein (CAP) of E. coli is a member of a class of sequence-specific DNA binding proteins that utilize a highly conserved alpha-helix-turn-alpha-helix structural motif to mediate interaction with DNA. This class of sequence-specifc DNA binding proteins has at least 75 members, including the homeo box proteins, which are involved in the regulation of eukaryotic development. Our research project regarding DNA-sequence recognition by CAP is directed at three long-range objectives: (i) identification of the individual amino acid-DNA contacts involved in DNA-sequence recognition by CAP, (ii) investigation of the chemistry and thermodynamics of specificity, and (iii) elucidation of rules--a "coded"--relating the amino acid sequence of the helix-turn-helix motif of a DNA binding protein to the nucleotide sequence of the specific DNA site. The experiments in this proposal have as specific aims the investigation of two examples of DNA-sequence recognition mediated through protein-base pair contacts, and the investigation of two examples of DNA-sequence recognition mediated through protein-phosphate contacts. DNA-sequence recognition mediated through protein-specific DNA binding protein. The approach to be employed emphasizes the use of single amino acid substitutions to identify and quantify important amino acid- DNA interactions. Single amino acid substitutions will be introduced into CAP using site-directed mutagenesis. The DNA- sequence recognition properties of the substituted CAP variants then will be investigated in in vitro binding assays using as ligands the consensus DNA site and a battery of non-consensus and chemically modified DNA sites; free energies of binding and free energies of specificity will be determine from the equilibrium binding data. The experimentally measured effects of the amino acid substitutions on the binding and specificity free energies will be interpreted in two fashions: (i) qualitatively, using computer-assisted molecule graphics and the known three- dimensional structures of CAP and B-DNA; and (ii) quantitatively, using molecular energy calculations. The results to be obtained will be relevant to understanding protein-nucleic acid interaction and its role in the regulation of gene expression. In addition, the results to be obtained will be relevant to understanding other examples of protein-ligand interaction, including: enzyme-substrate, receptor-hormone, and receptor-drug interactions.